Alzheimer's disease (AD) is the most common dementing illness of late life, and robs persons of vigorous activity and productivity in their later years. Future drug treatments may be capable of slowing the progression of the disease or even preventing its clinical appearance, and when such treatments become available, the ability to detect the illness in its earliest stages will be needed to take full advantage of them. The overall goal of this project is to test the use of structural magnetic resonance scanning and the tools of computational anatomy to distinguish nondemented subjects at elevated risk for developing dementia of the Alzheimer type (DAT) from age- and gender-matched comparison subjects. The primary strategy for identifying subjects at elevated risk for developing DAT will be having a parent with DAT; allelic status for the apolipoprotein E gene will be a secondary strategy. High resolution magnetic resonance (MR) scans will be collected from all subjects at entry into the study, and for subjects who are recruited during the first two years of the Project, repeat MR scans will be collected after three years. The specific aims of the project include a comparison of the structural characteristics of the hippocampus, parahippocampal gyrus (including the entorhinal cortex) and cingulate gyrus (anterior and posterior segments) in nondemented subjects with (N = 120) and without (n = 120) a parent with DAT. The total group of subjects will then be resorted according to the number of E4 alleles for apolipoprotein E, and a comparison of subjects with or without one or more E4 alleles will be made. If there are neuroanatomical differences in the groups of subjects with and without one or more E4 alleles, then the impact of the apoE4 allele on the effect of having a parent with DAT will be investigated. In addition, we will determine whether there are correlations between neuroanatomical measures that discriminate between subjects at elevated risk for developing DAT and comparison subjects and other predictive variables assessed by other Projects.